Computed Tomography Differentiates Forms of Adipose Masses in Dogs and Cats

A retrospective, cross-sectional study published in Veterinary Radiology & Ultrasound reported the computed tomography (CT) features distinguishing definitively diagnosed lipomas, infiltrative lipomas, and liposarcomas in a small sample of dogs and cats.

In domestic animals, adipose tumors are classified as either benign (lipoma, infiltrative lipoma, angiolipoma) or malignant (liposarcoma). Lipomas, which are common in dogs, contain well-differentiated adipocytes. Infiltrative lipomas are aggressive and invade nearby muscle and fascia. Angiolipomas, which are either infiltrative or noninfiltrative, contain well-differentiated blood vessels. Liposarcomas invade local tissue and have a high recurrence rate, but infrequently metastasize.

In veterinary medicine, CT is preferable over fine needle aspirates and small biopsies for differentiating between infiltrative lipomas and simple lipomas. However, information on CT features of adipose tumors in animals is limited. In addition, as authors noted, “a comparison between benign and malignant fatty masses based on CT characteristics in a larger group of animals has not been reported.”

For this study, authors selected 50 animals (46 dogs, 4 cats) that underwent CT evaluation and had definitively diagnosed lipomas, infiltrative lipomas, or liposarcomas. Data on each animal were collected from the databases of the Clinica Veterinaria dell’Orologio and the Royal (Dick) School of Veterinary Studies, University of Edinburgh. Definitive diagnoses of the adipose tumors were made based on histopathological examination using WHO criteria; neither cytoreductive surgeries nor incisional biopsies were performed before CT imaging.

Two board-certified veterinary radiologists, who were blinded to final diagnosis data prior to the study, retrospectively reviewed all CT images. For several adipose tumors, only pre- or postcontrast images were available for review. Reviewers evaluated each image for several criteria, including tumor volume and shape, tumor margins, type of intralesional hyperattenuating component (linear or nodular/globular/mass), regional lymphadenopathy, and potential metastasis.

Sixty adipose tumors were observed in the 50 animals; several of the dogs had multiple adipose tumors. Of these 60 tumors, 23 were lipomas (15 dogs, 3 cats), 20 were infiltrative lipomas (17 dogs, 1 cat), and 17 were liposarcomas (15 dogs). No angiolipomas were identified. Slightly over half (32/60) of adipose tumors were located within the trunk; others were in the thoracic or pelvic limbs (14/60), thorax or abdomen (10/60), or head-neck region (4/60).

Most of 23 lipomas were oval or round and had well-defined margins and homogeneous attenuation. Fat was the prevalent component in all lipomas. Ten lipomas contained intralesional hyperattenuation, which was linear (7/10) or nodular/globular/mass (3/10). Regional lymphadenopathy was present with four of the lipomas.

Many of the 20 infiltrative lipomas were irregularly shaped and had well-defined or mixed margins and homogeneous attenuation. Like the lipomas, fat was the prevalent component for all infiltrative lipomas. Seventeen infiltrative lipomas contained intralesional hyperattenuation, with all hyperattenuation being linear. Five cases of infiltrative lipomas had regional lymphadenopathy.

Most of the 17 liposarcomas were oval to round, with approximately half having well-defined margins. Unlike the benign adipose tumors, all liposarcomas had heterogeneous attenuation. Fat was the prevalent component in just over half of the liposarcomas; soft tissue was prevalent in the others. Intralesional hyperattenuation was primarily nodular/globular/mass, rather than linear. Regional lymphadenopathy was present with 10 of the liposarcoma cases. Four liposarcomas contained mineralization. One liposarcoma case had a potential metastatic lesion within the peritoneal cavity.

Authors identified significantly different CT features between lipomas and infiltrative lipomas, and between benign and malignant adipose tumors. Tumor shape, tumor margins, and presence and type of hyperattenuating component were significantly different between lipomas and infiltrative lipomas. Comparing benign and malignant adipose tumors, significantly different CT features included homogeneity, prevalent component, mineralization, regional lymphadenopathy, and type of hyperattenuating component.

Logistic regression analysis identified CT features with statistically significant odds ratios favoring a diagnosis of lipoma, infiltrative lipoma, or liposarcoma. Well-defined margins and an absence of hyperattenuating components favor a lipoma diagnosis, and an irregular shape favors an infiltrative lipoma diagnosis. Heterogenous attenuation, nodular/globular/mass hyperattenuating component, mineralization, regional lymphadenopathy, and prevalence of soft tissue favor a liposarcoma diagnosis.

Authors noted several study limitations, including the retrospective nature of the study, a small number of cases, and unavailability of corresponding pathological specimens. For future prospective studies, authors suggested comparing CT and histopathological findings “to further characterize CT features of malignancy for fat-containing tumors.”

​Dr. JoAnna Pendergrass received her doctorate in veterinary medicine from the Virginia-Maryland College of Veterinary Medicine. Following veterinary school, she completed a postdoctoral fellowship at Emory University’s Yerkes National Primate Research Center. Dr. Pendergrass is the founder and owner of JPen Communications, LLC.